OpenPipe Blog The latest updates, insights, and technical articles from OpenPipe https://openpipe.ai/blog en-us Wed, 19 Aug 2026 00:41:38 GMT hello@openpipe.ai (OpenPipe Team) hello@openpipe.ai (OpenPipe Team) <![CDATA[OpenPipe training and inference migrating to W&B]]> https://openpipe.ai/blog/openpipe-training-inference-migrating-to-wandb https://openpipe.ai/blog/openpipe-training-inference-migrating-to-wandb Mon, 18 May 2026 07:00:00 GMT Kyle and David Corbitt News <![CDATA[Serverless RL: Faster, Cheaper and More Flexible RL Training]]> https://openpipe.ai/blog/serverless-rl https://openpipe.ai/blog/serverless-rl Wed, 08 Oct 2025 07:00:00 GMT Kyle Corbitt Product <![CDATA[OpenPipe is Joining CoreWeave]]> https://openpipe.ai/blog/openpipe-coreweave https://openpipe.ai/blog/openpipe-coreweave Wed, 03 Sep 2025 07:00:00 GMT Kyle Corbitt News <![CDATA[📏RULER: Easy Mode for RL Rewards]]> https://openpipe.ai/blog/ruler https://openpipe.ai/blog/ruler Fri, 11 Jul 2025 07:00:00 GMT Kyle Corbitt Technical <![CDATA[Summary-RL]]> https://openpipe.ai/blog/summary-rl https://openpipe.ai/blog/summary-rl Wed, 25 Jun 2025 07:00:00 GMT Andie Jones Technical <![CDATA[Reward Hacking 101]]> https://openpipe.ai/blog/reward-hacking https://openpipe.ai/blog/reward-hacking Fri, 06 Jun 2025 07:00:00 GMT Kyle Corbitt Technical <![CDATA[ART·E: How We Built an Email Research Agent That Beats o3]]> https://openpipe.ai/blog/art-e-mail-agent https://openpipe.ai/blog/art-e-mail-agent Tue, 29 Apr 2025 07:00:00 GMT Kyle Corbitt Technical <![CDATA[ART Trainer: A New RL Trainer for Agents]]> https://openpipe.ai/blog/art-trainer-a-new-rl-trainer-for-agents https://openpipe.ai/blog/art-trainer-a-new-rl-trainer-for-agents Mon, 14 Apr 2025 07:00:00 GMT Kyle Corbitt News <![CDATA[pii-redact - SOTA PII Redaction on Your Laptop]]> https://openpipe.ai/blog/pii-redact https://openpipe.ai/blog/pii-redact Wed, 26 Mar 2025 07:00:00 GMT Andie Jones Technical <![CDATA[Using GRPO to Beat o1, o3-mini, and R1 on Temporal Clue]]> https://openpipe.ai/blog/using-grpo-to-beat-o1-o3-mini-and-r1-on-temporal-clue https://openpipe.ai/blog/using-grpo-to-beat-o1-o3-mini-and-r1-on-temporal-clue Thu, 06 Mar 2025 08:00:00 GMT Brad Hilton and Kyle Corbitt Technical <![CDATA[5 Deep Research Prompts that are Supercharging our Sales Strategy]]> https://openpipe.ai/blog/deep-research-prompts https://openpipe.ai/blog/deep-research-prompts Wed, 19 Feb 2025 08:00:00 GMT Daniel Bolus Basics <![CDATA[One Right Answer or Many? A Useful Distinction for Evaluating and Fine-Tuning LLMs]]> https://openpipe.ai/blog/deterministic-vs-freeform-tasks https://openpipe.ai/blog/deterministic-vs-freeform-tasks Tue, 14 Jan 2025 08:00:00 GMT Kyle Corbitt Technical <![CDATA[Analyzing OpenAI’s Reinforcement Fine-Tuning: Less Data, Better Results]]> https://openpipe.ai/blog/openai-rft https://openpipe.ai/blog/openai-rft Mon, 30 Dec 2024 08:00:00 GMT Kyle Corbitt Technical <![CDATA[Using Reinforcement Learning and $4.80 of GPU Time to Find the Best HN Post Ever (RLHF Part 1)]]> https://openpipe.ai/blog/hacker-news-rlhf-part-1 https://openpipe.ai/blog/hacker-news-rlhf-part-1 Mon, 28 Oct 2024 07:00:00 GMT Kyle Corbitt Technical <![CDATA[A Founder’s Guide to AI Fine-Tuning]]> https://openpipe.ai/blog/a-founder-s-guide-to-ai-fine-tuning https://openpipe.ai/blog/a-founder-s-guide-to-ai-fine-tuning Fri, 11 Oct 2024 07:00:00 GMT Kyle Corbitt Technical <![CDATA[Introducing Direct Preference Optimization (DPO) Support on OpenPipe]]> https://openpipe.ai/blog/announcing-dpo-support https://openpipe.ai/blog/announcing-dpo-support Tue, 01 Oct 2024 07:00:00 GMT Kyle Corbitt News <![CDATA[Fine-tuning Best Practices Chapter 2: Models]]> https://openpipe.ai/blog/fine-tuning-best-practices-chapter-2-models https://openpipe.ai/blog/fine-tuning-best-practices-chapter-2-models Wed, 28 Aug 2024 07:00:00 GMT Kyle Corbitt Technical <![CDATA[Fine-tuning Best Practices Series Introduction and Chapter 1: Training Data]]> https://openpipe.ai/blog/fine-tuning-best-practices-series-introduction-and-chapter-1-training-data https://openpipe.ai/blog/fine-tuning-best-practices-series-introduction-and-chapter-1-training-data Thu, 01 Aug 2024 07:00:00 GMT Kyle Corbitt Technical <![CDATA[Announcing Llama 3.1 and GPT-4o Mini fine-tuning through OpenPipe!]]> https://openpipe.ai/blog/llama-3-1-and-gpt-4o-mini-fine-tuning https://openpipe.ai/blog/llama-3-1-and-gpt-4o-mini-fine-tuning Wed, 24 Jul 2024 07:00:00 GMT Kyle Corbitt News <![CDATA[OpenPipe Mixture of Agents: Outperform GPT-4 at 1/25th the Cost]]> https://openpipe.ai/blog/mixture-of-agents https://openpipe.ai/blog/mixture-of-agents Thu, 20 Jun 2024 07:00:00 GMT Kyle Corbitt and Saumya Gandhi News <![CDATA[The Ten Commandments of Fine-Tuning in Prod (a Mastering LLMs Conference Talk)]]> https://openpipe.ai/blog/the-ten-commandments-of-fine-tuning-in-prod https://openpipe.ai/blog/the-ten-commandments-of-fine-tuning-in-prod Thu, 23 May 2024 07:00:00 GMT Kyle Corbitt Technical <![CDATA[What we've learned in 3 days of Llama 3]]> https://openpipe.ai/blog/what-we-ve-learned-in-3-days-of-llama-3 https://openpipe.ai/blog/what-we-ve-learned-in-3-days-of-llama-3 Sun, 21 Apr 2024 07:00:00 GMT Kyle Corbitt News <![CDATA[Fine-Tuning in a Nutshell]]> https://openpipe.ai/blog/a-non-technical-explanation-of-fine-tuning https://openpipe.ai/blog/a-non-technical-explanation-of-fine-tuning Thu, 28 Mar 2024 07:00:00 GMT David Corbitt Basics <![CDATA[We Raised $6.7M to Replace GPT-4 with Your Own Fine-Tuned Models]]> https://openpipe.ai/blog/announcing-6-7m-seed-raise https://openpipe.ai/blog/announcing-6-7m-seed-raise Mon, 25 Mar 2024 07:00:00 GMT Kyle Corbitt News <![CDATA[Mixtral Curious? Comparing Mistral 7B and Mixtral for fine-tuning]]> https://openpipe.ai/blog/comparing-mixtral-and-mistral-7b-for-fine-tuning https://openpipe.ai/blog/comparing-mixtral-and-mistral-7b-for-fine-tuning Thu, 29 Feb 2024 08:00:00 GMT Kyle Corbitt Comparison <![CDATA[S-LoRA: Serving Thousands of Models From One GPU for Fun and Profit]]> https://openpipe.ai/blog/s-lora https://openpipe.ai/blog/s-lora Wed, 17 Jan 2024 08:00:00 GMT Kyle Corbitt Technical <![CDATA[Axis Improves Generation Quality and Lowers Costs With Fine Tuning]]> https://openpipe.ai/blog/axis-case-study https://openpipe.ai/blog/axis-case-study Thu, 04 Jan 2024 08:00:00 GMT Kyle Corbitt Technical <![CDATA[Product Updates December 2023]]> https://openpipe.ai/blog/product-updates-december-2023 https://openpipe.ai/blog/product-updates-december-2023 Wed, 03 Jan 2024 08:00:00 GMT Kyle Corbitt News <![CDATA[How we built “Mistral 7B Fine-Tune Optimized,” the best 7B model for fine-tuning]]> https://openpipe.ai/blog/mistral-7b-fine-tune-optimized https://openpipe.ai/blog/mistral-7b-fine-tune-optimized Mon, 18 Dec 2023 08:00:00 GMT Kyle Corbitt Comparison <![CDATA[Announcing Automatic Evals for Fine-tuned Models]]> https://openpipe.ai/blog/announcing-automatic-evals-for-fine-tuned-models https://openpipe.ai/blog/announcing-automatic-evals-for-fine-tuned-models Fri, 01 Dec 2023 08:00:00 GMT Kyle Corbitt News <![CDATA[Is AI the Next Crypto? Insights from 2M HN comments]]> https://openpipe.ai/blog/hn-ai-crypto https://openpipe.ai/blog/hn-ai-crypto Wed, 08 Nov 2023 08:00:00 GMT Kyle Corbitt Technical <![CDATA[Llama 2 vs Mistral: Believe the Hype]]> https://openpipe.ai/blog/llama-2-vs-mistral-believe-the-hype https://openpipe.ai/blog/llama-2-vs-mistral-believe-the-hype Sun, 05 Nov 2023 07:00:00 GMT Kyle Corbitt Technical <![CDATA[Fine-tune your own Llama 2 to replace GPT-3.5/4]]> https://openpipe.ai/blog/fine-tune-your-own-llama-2-to-replace-gpt-3-5-4 https://openpipe.ai/blog/fine-tune-your-own-llama-2-to-replace-gpt-3-5-4 Tue, 12 Sep 2023 07:00:00 GMT Kyle Corbitt Technical <![CDATA[From Prompts to Models]]> https://openpipe.ai/blog/from-prompts-to-models https://openpipe.ai/blog/from-prompts-to-models Mon, 28 Aug 2023 07:00:00 GMT Kyle Corbitt and David Corbitt Comparison